Enroll in the course

Machine Learning for Physicists

For students who want to use machine learning in academic and research fields
Course Registration Deadline: September 20, 2021

Join the webinar to learn more: September 15 at 12:00pm MSK

Offered by

Group 76
Group 78

With the IT sector expanding rapidly around the world, there has never been a better time to take a machine learning course. The Machine Learning for Physicists course will provide you with the knowledge and practical skills you need to further your career or research activities.

The course is available for undergraduate students gearing up for their 4th year of study, designed for programmer beginners. You will gain introductory knowledge that will help you grasp common sense concepts and advanced techniques on machine learning.

You will receive a theoretical and practical introduction to this new field and will be able to apply acquired knowledge to solve your own problems. Topics will range from decision trees to deep learning and simulation-based inference and will be covered with concrete examples and hands-on tutorials.

Get ready to gain key insights on the following modules: Intro into Deep Learning, Supervised Deep Learning Models, Unsupervised Deep Learning Models, Advanced Learning.

Join the Webinar

Join us September 15 at 12:00pm MSK for an online webinar and discover why Machine Learning for Physicists is a good fit for you.

Meet the speakers:

  • Dmitry Livanov, Rector of MIPT
  • Stanislav Protassov, Technology President and Co-founder, Acronis
  • Kostya Novoselov, Nobel Prize winner (2010), Chairman of the SIT Strategic Advisory Board, Professor of Physics Center for Advanced Materials, National University of Singapore
  • Andrey Ustyuzhanin, AI/MI expert consultant at SIT, the director of the Laboratory of Methods for Big Data Analysis at the Higher School of Economics
  • Ilya Shimchik, Acronis SIT Autonomous Team Principal
Group 18
Course Skills
Dynamics, ode, machine learning, probability, data-driven, optimization, linear algebra, calculus, gradient descent, Newton method, regression, programming, quantum theory, complex number 
Group 30
For Computer Science Beginners
The course is designed for the 4th year students.
Group 15
Course Details
1 year, 2 semesters, 120 hours. Online course on automated testing with monthly Q&A live sessions from instructors.
The course is taught in English


Andrey Ustyuzhanin
Dr. Andrey Ustyuzhanin is an AI/MI expert consultant at SIT, the director of the Laboratory of Methods for Big Data Analysis at the Higher School of Economics. His team participates in several international collaborations: LHCb, SHiP. The primary focus of his research is the design and application of Machine Learning methods to improve fundamental understanding of our world principles.
Mikhail Hushchyn
Researcher at the National Research University Higher School of Economics, Moscow; PhD in mathematical modeling, numerical methods and complexes of programs.
Alena Zarodnyuk
Researcher at the National Research University Higher School of Economics, Moscow; PhD in Physics and Mathematics.
Leonid Gremyachikh
Ph.D. student and Junior Researcher at the Faculty of Computer Science in the HSE University. Among his scientific interests are such areas as Artificial Intelligence, Data Science, Deep Learning, Reinforcement Learning and Computer Vision.

Course Overview

This course, designed for those with little to no programming experience, will provide students with a non-conventional introduction into deep machine learning, understanding of both supervised and unsupervised deep machine learning models and advanced machine learning over the course of 2 semesters (120 hours of lessons).

Get acquainted with the main machine learning algorithms used in today’s academic and business research across different fields: deep learning, convolutional neural networks, computer vision, time series, generative networks, autoencoders, neurodiffs, and optimization methods.

Topic 1: Intro into Deep Learning

Begin with the basics of Machine Learning and Deep Learning. Get a grasp on the key terms, problems and ML algorithms of supervised and unsupervised learning. Gain insights into gradient optimization methods and the architecture of artificial, convolution and recurrent neural networks. Find out how to train neural networks and assess their outputs. Discover what neural network under- and overfitting is and how to avoid it.

Topic 2: Supervised Deep Learning Models

Get an intro into computer vision, time series and graph structures. Learn how to use specialized computer vision libraries, remove image noises, restore images, detect objects in images and more. Find out what time series is, how to filter signals, the value of using Fourier series and how to forecast chaotic time series.

Topic 3. Unsupervised Deep Learning Models

Zoom in on machine learning algorithms based on neural networks with unsupervised learning. Get acquainted with AutoEncoders and generative models (GANs, Flows) to fill in the blanks, translate image-to-image and speed up synthetic data output by replacing the simulator with a neural network.

Topic 4. Advanced Learning

Get an overview of advanced optimization methods, neural ordinary differential equations (NeuralODE) and automatic relevance determination. Explore modern approaches to optimization, including particle swarm optimization (PSO), evolutionary algorithm (EA), Bayesian optimization and more. Learn more about Neural ODE and their relation to optimal control and backward problem solution.

Enrollment Details

MIPT students from the Department of General and Applied Physics can choose one course between: “Machine Learning for Physicists“, “Oscillations and Waves” and “Electronic methods of physical research".

Students from other Phystech Schools can include “Machine Learning for Physicists“ to their individual plan.

The course is designed for students entering their 4th year of study.

Register to attend the webinar and to receive enrollment details:

Logo Alemira

About SIT Alemira

SIT Alemira provides a complete digital ecosystem for education and learning. Through an AI-powered authoring platform, SIT Alemira supports universities like MIPT to create adaptive content for a more collaborative research design approach. Advised by leading education experts, SIT Alemira leverages Active Learning technologies to deliver hands-on learning experiences and complex results, enhancing the online learning experience.

About SIT

Schaffhausen Institute of Technology (SIT), located in Schaffhausen, Switzerland, is an international institution founded by entrepreneurs, led by scientists and advanced by world-class researchers. Interdisciplinary by design, SIT comprises a unique ecosystem, including that of SIT Alemira, where the world’s leading experts in Computer Science, Physics and Business come together to find innovative solutions to global challenges through transformative technological advances.